Intelligent building method for parameterized three-dimensional power grid frame

Abstract

The invention provides an intelligent building method for a parameterized three-dimensional power grid frame, and relates to the technical field of electric power, and the method comprises the steps of first collection, second collection, third collection, model construction, demand prediction, power grid division, first setting, architecture construction and the like. The charging demand fluctuation condition of the electric vehicle can be dynamically monitored, the power grid is divided by adopting the clustering algorithm according to the charging demands and regional characteristics of different regions, the load center is set based on the power grid division result, and the power grid architecture is constructed. The constructed power grid architecture can better cope with the conditions of power load fluctuation and power load regional difference caused by popularization of electric vehicles, the dynamic response capability of the power grid is enhanced, and the stability of the power grid is improved.

Description

Technical Field

[0001] The present application relates to the field of electric power technology, and in particular to a method for intelligently constructing a parameterized three-dimensional power grid framework. Background Art

[0002] With the acceleration of urbanization and the rapid development of the electric vehicle (EV) industry, the traditional power grid architecture faces huge challenges. The widespread popularity of electric vehicles has brought about a surge in charging demand, especially in high-density urban areas. The load of the power grid is highly time-varying and regional, and the original power grid architecture is often unable to cope with the fluctuations in charging load.

[0003] The Chinese invention patent with application publication date of March 30, 2022 and publication number CN115238943A provides an active distribution network planning method, which obtains the basic data of the active distribution network; based on the acquired data, constructs an active distribution network planning model with the minimum annual comprehensive cost as the objective function, taking into account the uncertainty of renewable energy backup and the delay of electric vehicle backup; linearizes the nonlinear active distribution network planning model, and uses a mixed integer linear programming algorithm to solve the linearized active distribution network planning model to obtain a planning scheme for the active distribution network.

[0004] For the above technical solution, the goal is to minimize the annual comprehensive cost, taking into account the reserve uncertainty of renewable energy and electric vehicles, and solving the distribution network planning problem through a mixed integer linear programming algorithm. However, the above solution focuses on the statistics and planning of historical data, while ignoring the dynamic impact of dynamic factors such as holidays and traffic flow on the charging load of electric vehicles. It cannot respond in time to the time-varying and regional impact of the widespread use of electric vehicles on the load of the power grid, which leads to local overload or unreasonable allocation of power grid resources. Summary of the invention

[0005] In order to better cope with the load fluctuations and regional differences caused by the popularization of electric vehicles, enhance the dynamic response of the power grid, and improve the stability and efficiency of the power grid, the present application provides a parameterized three-dimensional power grid intelligent construction method.

[0006] In the first aspect, the present application provides a method for intelligently constructing a parameterized three-dimensional power grid framework, which adopts the following technical solutions: A parameterized three-dimensional power grid intelligent construction method comprises the following steps: First collection: collect geographical data, power grid equipment data and regional characteristic data of each region according to the geographic information system; the power grid equipment data includes the location data of charging piles; Second collection: collect the first real-time data of each area, including holiday conditions, regional traffic flow and car owners' travel habits; Third collection: collecting the first historical data of each area, the first historical data includes the historical charging data of each charging pile, the historical regional traffic flow, the historical holiday conditions and the historical travel habits of car owners; Model construction: Build a charging demand prediction model; Demand forecasting: inputting the first real-time data into the charging demand forecasting model to obtain charging demand forecasting data of each charging pile; Grid division: Cluster the charging piles based on their location data, charging demand forecast data and regional feature data. According to the clustering results, several sub-areas are obtained. The sum of the charging demand forecast data of the charging piles in each sub-area is used as the forecast load of each sub-area. The first setting is: divide the sub-areas according to the preset grid size, divide the predicted load of each sub-area based on the area of ​​each grid, obtain the predicted load of each grid, obtain the geometric center coordinates of each grid, use the weighted centroid method to calculate the coordinates of the load center of each sub-area, and use the predicted load of each sub-area as the load capacity of the load center; Architecture construction: Based on the load capacity of the load centers in each sub-region, the load centers in each sub-region are interconnected to build the power grid architecture.

[0007] By adopting the above technical solution, a charging demand prediction model is constructed to predict the charging demand of each charging pile in different time periods, which can accurately predict future charging demand, identify charging peak and valley periods in advance, help to reasonably dispatch power resources, and reduce the overload of the power grid due to sudden load fluctuations. In addition, based on the location data of each charging pile, the charging demand prediction data of each time period and the regional characteristic data, the clustering algorithm is used to divide the power grid into regions. According to the location of the charging pile and the corresponding charging demand prediction data, it is possible to identify areas with high and low charging demand and areas with similar charging demand time fluctuations, so as to allocate power resources in a targeted manner, reduce the load overload and idle resources in each area, help to plan the power grid capacity more reasonably, thereby optimizing the allocation of power resources and improving the adaptability of the power grid to load changes. Moreover, based on the divided power grid, the load center is set, and the appropriate electrical equipment and transmission lines are selected based on the set load center to build the power grid architecture, which can accurately grasp the predicted load changes in each sub-region, adjust the power distribution in time, optimize the load distribution of the power grid, and improve the stability and security of the power grid.

[0008] Optionally, after executing the third acquisition step and before executing the model building step, the method further includes: First area division: Based on the regional characteristic data of each area, a clustering algorithm is used to perform the first area division on each area to obtain a new area grouping result, which is recorded as the first traffic area; First data matching: mapping the historical vehicle owner travel habits and historical regional traffic flow to each first traffic area, obtaining new historical vehicle owner travel habits and new historical regional traffic flow in each first traffic area, which are recorded as first historical vehicle owner travel data and first historical traffic flow data respectively; Second data matching: using historical holiday conditions to divide the first historical vehicle owner travel data and the first historical traffic flow data, respectively, to obtain non-holiday historical vehicle owner travel data, holiday historical vehicle owner travel data, non-holiday historical traffic flow data, and holiday historical traffic flow data, and record the non-holiday historical vehicle owner travel data and non-holiday historical traffic flow data as the first traffic data, and record the holiday historical vehicle owner travel data and holiday historical traffic flow data as the second traffic data; Calculate the transfer probability: based on the first traffic data and the second traffic data, use the Markov chain method to calculate the first vehicle transfer probability and the second vehicle transfer probability of each first traffic area, and record the first vehicle transfer probability and the second vehicle transfer probability of the first traffic area as the comprehensive transfer probability of the first traffic area; Traffic flow dynamic modeling: collect the comprehensive transfer probabilities of each first traffic area and build a traffic flow dynamic model; Obtaining traffic flow distribution: Inputting regional traffic flow into the traffic flow dynamic model, obtaining predicted traffic flow distribution data for each region, and using the predicted traffic flow distribution data as the new regional traffic flow.

[0009] By adopting the above technical solutions, a dynamic model of vehicle flow is constructed. Combined with the comprehensive transfer probability, it is possible to dynamically simulate the distribution of vehicle flow between regions. By predicting future traffic flow, the prediction range of the charging demand prediction model can be extended to several days or even weeks, which is suitable for charging demand planning and grid load scheduling on a longer time scale. It helps to identify the peak charging period of charging piles in advance, schedule the grid load in advance, and reduce the overload of local grids due to concentrated load. In addition, the dynamic model of vehicle flow can reflect the changes in real-time and long-term traffic flow trends, improve the adaptability of the charging demand prediction model to complex scenarios, enhance the accuracy and adaptability of charging demand prediction and grid load management, and improve the stability and efficiency of the grid.

[0010] Optionally, after performing the first area division step and before performing the first data matching step, the method further includes: Constructing a traffic network diagram: mapping the travel habits of car owners and regional traffic flow in each first traffic area, obtaining the travel habits of car owners and regional traffic flow in each first traffic area, and constructing a traffic network diagram with each first traffic area as a node and the travel habits of car owners in each first traffic area and the corresponding regional traffic flow as an edge; Second area division: Based on the traffic network graph, a graph division algorithm is used to perform a second area division on each first traffic area, and the result of the second area division is used as the new first traffic area.

[0011] By adopting the above technical solution, the traffic network graph and graph division algorithm are introduced in the process of regional division, which makes the regional division more reasonable, helps the traffic flow dynamic model to adapt to the traffic network characteristics of different cities or regions, improves the scalability of the traffic flow dynamic model, helps the traffic flow dynamic model to more accurately reflect the traffic flow inside and outside each region, and improves the accuracy of the traffic flow dynamic model prediction.

[0012] Optionally, after executing the step of calculating the transition probability and before executing the step of dynamic modeling of vehicle flow, the method further includes: Fourth data collection: collecting historical condition data corresponding to the first historical vehicle owner travel data and the first historical traffic flow data, the historical condition data including historical weather conditions and historical electricity prices; Defining a Bayesian network: defining Bayesian network variables and causal relationships between variables based on the historical travel habits of vehicle owners in each first traffic area and the corresponding historical condition data; the Bayesian network variables include child variables and parent variables; Construct a conditional probability table: fully arrange the parent variable values ​​of the child variables to obtain several condition combinations, calculate the probability value of each condition combination based on historical condition data and historical vehicle owner travel habits, and collect the probability values ​​of each condition combination to obtain a conditional probability table; Optimize the transfer probability: based on the historical condition data of each first traffic area, obtain the condition data corresponding to the first traffic data and the second traffic data, query the condition probability table according to the obtained condition data, obtain the condition probability, use the condition probability to optimize the comprehensive transfer probability of the first traffic area, and use the optimized comprehensive transfer probability as the new comprehensive transfer probability; After executing the step of dynamic modeling of vehicle flow, and before executing the step of obtaining vehicle flow distribution, the method further includes: Fifth, data collection: collect real-time condition data, including weather conditions and electricity prices; Data labeling: Use timestamps to match weather conditions, electricity prices, and regional traffic flows, label the regional traffic flows with weather conditions and electricity prices, and use the labeled regional traffic flows as new regional traffic flows.

[0013] By adopting the above technical solution, introducing Bayesian networks and conditional probability tables, and considering the impact of external factors on car owner behavior and traffic flow, the prediction accuracy and adaptability of the vehicle flow dynamic model are improved, and the accuracy of traffic flow prediction is improved. It not only improves the accuracy and real-time performance of the charging demand prediction model, but also helps the power grid and charging pile operators to dispatch resources more efficiently, optimize services, and reduce the imbalance between supply and demand caused by sudden changes in conditions, so that the power grid can more effectively cope with complex and changeable traffic environments.

[0014] Optionally, after executing the step of building the model and before executing the step of demand forecasting, the method further includes: Constructing a sample training set: constructing a sample training set based on the first historical data; Encoder training: The sample training set is input into the charging demand prediction model, and the Transformer encoder encodes the sample data in the sample training set; Decoder prediction: The Transformer decoder decodes and predicts the encoded sample data to obtain charging demand prediction data; Iterative optimization: define a mean square error loss function, iteratively optimize the charging demand prediction model by minimizing the mean square error loss function, and use the optimized charging demand prediction model as the new charging demand prediction model.

[0015] By adopting the above technical solutions, the self-attention mechanism of the Transformer model can capture the long-distance dependencies in the sample data, identify the long-term trends and periodic fluctuations in the sample data, and handle the differences in charging demand at different charging piles in different periods of time, further improving the adaptability of the charging demand prediction model.

[0016] Optionally, after executing the step of dividing the power grid and before executing the step of first setting, the method further includes: First judgment: determine whether there is a charging pile in the sub-area: If yes, then execute the second judgment step; If not, then execute the first setting step; Second judgment: judge whether the number of charging piles in the sub-area is greater than 1: If yes, the load weight of each charging pile is calculated according to the charging demand forecast data of each charging pile, the coordinates of the load center of the current sub-area are calculated by the weighted centroid method according to the positions of all charging piles in the current sub-area and the corresponding load weights, the predicted load of the current sub-area is used as the load capacity of the load center, and the step of architecture construction is executed; If not, the coordinates of the charging pile are used as the load center of the current sub-area, the predicted load of the current sub-area is used as the load capacity of the load center, and the architecture construction step is executed.

[0017] By adopting the above technical solution, when there are no charging piles in the sub-area, the load center is directly set, which simplifies the processing flow; when there are multiple charging piles, the load weight is calculated through charging data, and the load weight and position of each charging pile are considered. The load center is set using the weighted centroid method, so that the load center is closer to the actual charging demand distribution, ensuring that the location of the load center can best represent the charging demand distribution in the area, improving the rationality, flexibility and accuracy of the first setting, helping to optimize the allocation of power resources, reduce operating costs, make power transmission more efficient, reduce unnecessary power waste, and improve the power supply efficiency and stability of the power grid.

[0018] Optionally, after executing the step of dividing the power grid and before executing the step of first setting, the method further includes: Calculate load density: According to the geographical data of each sub-region, obtain the area of ​​each sub-region, calculate the load density of each sub-region in each period according to the area of ​​each sub-region and the predicted load of each charging pile in each sub-region in each period, aggregate the load density of each sub-region in each period, and obtain the load density characteristic vector of each sub-region; Secondary division of power grid: clustering algorithm is used to cluster each sub-region into different preliminary sub-regions using the load density feature vector of each sub-region, and the preliminary sub-regions are aggregated to obtain a preliminary region set; Constructing a regional connection graph: Based on the power grid equipment data of each region, each preliminary sub-region is used as a graph node, and the transmission lines between the preliminary sub-regions are used as edges to construct a regional connection graph; Region optimization: Based on the region connection graph, the minimum cut algorithm is used for optimization, and the optimized preliminary sub-regions are used as new sub-regions.

[0019] By adopting the above technical solution, in the process of power grid division, by calculating the load density characteristic vector of the sub-region, the influence of the regional area and the charging load distribution is comprehensively considered, so that the sub-region division result is more in line with the actual needs. In addition, based on the obtained preliminary regional set, the regional connection diagram is constructed, and the division of sub-regions is further optimized through the minimum cut algorithm, so that the power grid can adapt to the demand changes more flexibly when facing the fluctuation of charging demand in different regions, thereby improving the adaptability of the power grid. Moreover, optimization based on the consideration of load density and transmission network structure is conducive to the formulation of a more scientific long-term development plan for the power grid.

[0020] Optionally, after executing the step of building the architecture, the method further includes: Third judgment: obtain the actual charging load of each sub-area, calculate the difference between the actual charging load and the predicted load of each sub-area, record it as the first difference, and judge whether the first difference of the sub-area is higher than the preset first load threshold: If yes, the sub-region where the first difference is higher than the first load threshold is recorded as an overload region, and the fourth determination step is executed; If not, no action will be taken; Fourth judgment: record the absolute value of the first difference of the remaining sub-areas except the overload area as the second difference, and judge whether the second difference of the remaining sub-areas except the overload area is higher than the preset second load threshold: If yes, the sub-area where the second difference is higher than the second load threshold is recorded as an idle area, the difference between the second difference of the idle area and the preset third load threshold is calculated and recorded as the deployed load, the deployed load of the idle area is deployed to the overloaded area, and the charging power of the charging piles in the overloaded area is reduced; If not, the charging power of the charging piles in the overload area is reduced.

[0021] By adopting the above technical solution, setting multiple load thresholds and performing difference judgment, it is possible to handle different load scenarios and accurately control the risks of load overload and excessive load. By dynamically adjusting the charging power and load distribution, long-term overload operation is reduced, which helps to extend the service life of charging piles and power grid equipment and reduce equipment maintenance costs.

[0022] In summary, the present application includes at least one of the following beneficial technical effects: 1. Construct a charging demand prediction model to predict the charging demand of each charging pile in different time periods, which can accurately predict future charging demand and identify charging peak and valley periods in advance, which is helpful to realize reasonable dispatch of power resources and reduce the overload of the power grid due to sudden load fluctuations. In addition, based on the location data of each charging pile, the charging demand prediction data of each time period and the regional characteristic data, the clustering algorithm is used to divide the power grid into regions. According to the location of the charging pile and the corresponding charging demand prediction data, it is possible to identify areas with high and low charging demand and areas with similar charging demand time fluctuations, so as to allocate power resources in a targeted manner, reduce the overload and idle resources in each area, and help to plan the power grid capacity more reasonably, thereby optimizing the allocation of power resources and improving the adaptability of the power grid to load changes. Moreover, based on the divided power grid, the load center is set, and the appropriate electrical equipment and transmission lines are selected based on the set load center to build the power grid architecture, which can accurately grasp the predicted load changes in each sub-region, adjust the power distribution in time, optimize the load distribution of the power grid, and improve the stability and security of the power grid.

[0023] 2. Constructing a dynamic model of vehicle flow, combined with comprehensive transfer probability, can dynamically simulate the distribution of vehicle flow between regions. By predicting future traffic flow, the prediction range of the charging demand prediction model can be extended to several days or even weeks, which is suitable for longer-term charging demand planning and grid load scheduling. It helps to identify the peak charging period of charging piles in advance, and schedule the grid load in advance, reducing the overload of local grids due to concentrated load. Moreover, the dynamic model of vehicle flow can reflect the changes in real-time and long-term traffic flow trends, improve the adaptability of the charging demand prediction model to complex scenarios, enhance the accuracy and adaptability of charging demand prediction and grid load management, and improve the stability and efficiency of the grid.

[0024] 3. In the process of grid division, by calculating the load density characteristic vector of the sub-region, the influence of the regional area and the charging load distribution is comprehensively considered to make the sub-region division result more in line with the actual needs. In addition, based on the constructed regional connection map and the obtained preliminary regional set, the division of sub-regions is further optimized through the minimum cut algorithm, so that the grid can more flexibly adapt to the changes in demand when facing the fluctuation of charging demand in different regions, thus improving the adaptability of the grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 is a flow chart of Example 1 of the present application; Figure 2 It is a flowchart of S41 model training in Example 1 of the present application; Figure 3 This is a flow chart of S31 traffic flow prediction in Example 2 of the present application; Figure 4 It is a flowchart of S317 conditional probability optimization of Example 2 of the present application. DETAILED DESCRIPTION

[0026] The following combination Figures 1 to 4 This application is described in further detail.

[0027] Embodiment 1: This embodiment discloses a method for intelligently constructing a parameterized three-dimensional power grid. Figure 1As shown, the construction method includes: collecting geographic data, power grid equipment data, regional characteristic data, first real-time data and first historical data of each region, and constructing a charging demand prediction model based on the Transformer model; inputting the first real-time data into the charging demand prediction model to obtain charging demand prediction data of each charging pile, clustering the charging piles based on the location data of each charging pile, the charging demand prediction data of each charging pile and the regional characteristic data, and obtaining a number of sub-regions according to the clustering results, taking the sum of the charging demand prediction data of the charging piles in each sub-region as the predicted load of each sub-region, dividing each sub-region into grids, calculating the coordinates of the load center of each sub-region based on the geometric center coordinates of the grid, taking the predicted load of each sub-region as the load capacity of the load center, and connecting the load centers of each sub-region to each other based on the load capacity of the load center of each sub-region to construct a power grid architecture. This embodiment includes the following steps: S1 First collection: Collect geographic data, power grid equipment data and regional characteristic data of each region based on the geographic information system; the power grid equipment data includes charging pile location data.

[0028] The geographic data of each region is collected using a geographic information system, and the geographic data includes regional boundaries, regional terrain, regional coordinates, regional area, and regional shape. The power grid equipment data includes the path, capacity, and electrical characteristics of the transmission line. The power grid equipment data also includes electrical equipment, and the electrical equipment includes the location data of charging piles. The regional characteristic data includes the population density in the region and the building functions in each region. The building functions include commercial areas, industrial areas, office areas, and residential areas.

[0029] In this embodiment, the regions are administrative regions of the region.

[0030] S2 Second Collection: Collect the first real-time data of each area, the first real-time data includes holiday conditions, regional traffic flow and car owners' travel habits.

[0031] Regional traffic flow includes electric vehicle traffic flow data in the region and electric vehicle traffic flow change data in each time period. Regional traffic flow can be obtained through the traffic management platform. Holidays include working days and non-working days. Non-working days include statutory holidays (including Spring Festival, Qingming Festival, Labor Day, Dragon Boat Festival, Mid-Autumn Festival, National Day, etc.), mobile holidays (including Children's Day, etc.), Saturdays and Sundays. Working days generally refer to Monday to Friday. In some cases, there are overlaps between holidays, between holidays and Saturdays and Sundays, and between holidays and working days. At this time, the dates that overlap with holidays are marked as corresponding holidays. For example, the Dragon Boat Festival includes Friday, Saturday and Sunday, collectively referred to as the Dragon Boat Festival. Holidays are imported through the electronic calendar system. The travel habits of car owners include the travel frequency, travel time and travel route of electric vehicles, which can be obtained through the vehicle-mounted GPS device or the traffic management platform.

[0032] S3 third collection: collect the first historical data of each area, the first historical data includes the historical charging data of each charging pile, the historical regional traffic flow, the historical holiday conditions and the historical travel habits of the car owners.

[0033] The historical charging data of each charging pile includes the historical charging pile status, historical charging power and corresponding historical charging time. The historical regional traffic flow includes the historical traffic flow data of electric vehicles in the region and the historical traffic flow change data of electric vehicles in various time periods. The historical travel habits of car owners include the travel frequency, travel time and travel route of electric vehicles, which can be obtained through the vehicle GPS device or traffic management platform. The historical holidays are imported through the electronic calendar system.

[0034] In this embodiment, after the above data is collected, it is necessary to pre-process the data. The steps of data pre-processing include data cleaning and data normalization.

[0035] Data cleaning: performing operations including but not limited to denoising, processing missing values, and removing outliers on the collected geographic data of each region, power grid equipment data, regional characteristic data, first real-time data, and first historical data.

[0036] Reduce noise interference in data through smoothing, frequency domain filtering and outlier detection and denoising. Handle missing values ​​through interpolation, mean filling, nearest neighbor interpolation, regression filling and deletion. Identify outliers and remove outliers through visualization methods such as box plots, scatter plots and histograms.

[0037] Data normalization: normalize the collected geographic data of each region, power grid equipment data, regional characteristic data, first real-time data and first historical data.

[0038] S4 Model construction: Based on the Transformer model, a charging demand prediction model is constructed, and the charging demand prediction model includes an encoder and a decoder. In this embodiment, the charging demand prediction model is provided with a self-attention mechanism.

[0039] S41 model training: including S411 first data processing, S412 second data processing, S413 building a sample training set, S414 encoder training, S415 decoder prediction and S416 iterative optimization, such as Figure 2 shown.

[0040] S411 first data processing: divide the charging pile into different charging time periods based on the historical charging time in the historical charging data of each charging pile, and use the historical charging time, divided charging time periods and historical charging power to calculate the historical charging demand of each charging pile in different time periods, which is recorded as the first historical charging data.

[0041] S412 second data processing: Match the historical holiday conditions with the historical regional traffic flow through time stamps, then label the historical regional traffic flow with the corresponding historical holiday conditions, and record the labeled regional traffic flow as the first historical labeled data.

[0042] S413 constructs a sample training set: extracts the time features of the first marked historical data and the first historical charging data, takes the data at the same time as a group of input data, records them as a sample data, and collects all the sample data to obtain a sample training set.

[0043] S414 encoder training: The sample data in the sample training set is input into the charging demand prediction model. The Transformer encoder converts the input sample data into context-related feature representations through multiple layers of self-attention layers and feedforward neural networks, and passes the output of the Transformer encoder (i.e., the context-related feature representation) as input to the Transformer decoder.

[0044] S415 decoder prediction: The Transformer decoder decodes the context-related feature representation output by the Transformer encoder to predict future charging demand and obtain charging demand prediction data.

[0045] S416 iterative optimization: define a mean square error loss function, use the defined mean square error loss function to calculate the loss value of the charging demand prediction model, use the back propagation algorithm to calculate the model parameter gradient of the charging demand prediction model, use the gradient descent method to update the model parameters of the charging demand prediction model, until the preset number of iterative training times is completed or the calculated loss value of the charging demand prediction model no longer decreases, the model training is completed, and the optimized charging demand prediction model is obtained, and the optimized charging demand prediction model is used as the new charging demand prediction model.

[0046] The mean square error loss function formula is: .

[0047] in, represents the mean square error loss function of the charging demand prediction model, represents the number of samples in the sample training set, Indicates The actual charging demand of samples, The model predicts the Charging demand forecast data for samples.

[0048] S42 first marking: performing time-series matching of holiday conditions and regional traffic flow through timestamps, marking the regional traffic flow with labels corresponding to the holiday conditions, and recording the marked regional traffic flow as first marking data.

[0049] S5 demand prediction: input the first marked data into the charging demand prediction model, perform model reasoning, and obtain the charging demand prediction data of each charging pile in each time period.

[0050] S6 grid division: Based on the location data of each charging pile, the charging demand forecast data for each time period and the regional characteristic data, the charging piles are clustered using weighted distance clustering (K-means clustering or hierarchical clustering) or DBSCAN clustering algorithm. Each clustering result is taken as a sub-region, and each unclustered region is also taken as a sub-region to obtain several sub-regions. The sum of the charging demand forecast data of the charging piles in each sub-region is taken as the predicted load of each sub-region.

[0051] In this embodiment, the K-means clustering method is used to divide the power grid: Each region is recorded as an original region, and the geographical location and regional characteristic data of each original region are obtained by using a geographic information system, and the geographical location and regional characteristic data of each original region are standardized.

[0052] The time characteristics of the charging demand forecast data of each charging pile in each time period are extracted to obtain the charging demand time characteristic data of each charging pile every day.

[0053] The building functions in the regional characteristic data of each original area are uniquely encoded to obtain the encoded building functions. A linear regression model is constructed, and the charging demand forecast data of charging piles is selected as the target variable. The relationship between the daily charging demand forecast data of charging piles and the population density and building functions in the regional characteristic data is analyzed to obtain the regression weights of population density and each building function.

[0054] The formula for the linear regression model is: .

[0055] .

[0056] in, Represents the charging demand forecast data of the charging pile, represents the model intercept, represents the regression weight of population density, The building function is The regression weights, The building function is The regression weights, The building function is The regression weights, The building function is The regression weights, represents the total regression weight of building function, represents the error term.

[0057] in, =1.

[0058] Obtain the location data of each charging pile, and calculate the geographical distance between each pair of charging piles using the Haversian formula or Euclidean distance.

[0059] In this embodiment, the Euclidean distance is used to calculate the geographical distance between each pair of charging piles. The Euclidean distance formula is: .

[0060] in, Indicates charging pile With charging pile The geographical distance between Indicates charging pile Longitude, Indicates charging pile The longitude, Indicates charging pile The latitude, Indicates charging pile Latitude.

[0061] Calculate the regional characteristic weighted distance of each pair of charging piles. The regional characteristic weighted distance formula is: .

[0062] in, Indicates the area where the charging pile is located The area where the charging station is located The weighted distance between regional features, represents the regression weight of population density, Indicates the area where the charging pile is located population density, Indicates the area where the charging pile is located population density, represents the total regression weight of building function, Indicates the area where the charging pile is located The building function Indicates the area where the charging pile is located The building functions include =1.

[0063] Based on the geographical distance and regional characteristic weighted distance between each pair of charging piles, the weighted Euclidean distance between each charging pile is calculated.

[0064] The formula for weighted Euclidean distance is: .

[0065] in, Indicates charging pile With charging pile The weighted Euclidean distance between .

[0066] The K-means clustering algorithm is adopted to cluster the areas where the charging piles are located using the calculated weighted Euclidean distance. Each clustering result is taken as a sub-area, and each unclustered area is taken as a sub-area to obtain several sub-areas. The sum of the charging demand forecast data of all charging piles in each sub-area is taken as the predicted load of each sub-area.

[0067] S7 First setting: Divide each sub-region according to the preset grid size, divide each sub-region into multiple grids, and form a grid structure, where each grid represents a small geographical unit. Divide the predicted load of each sub-region according to the area of ​​each grid to obtain the predicted load of each grid in the sub-region. Use the geographic information system to obtain the geometric center coordinates of each grid in the sub-region, and use the first weighted centroid formula to calculate the coordinates of the load center of each sub-region. Use the predicted load of each sub-region as the load capacity of the load center.

[0068] In this embodiment, the grid size can be set according to actual needs.

[0069] The first weighted centroid formula is: ; .

[0070] in, Indicates the longitude of the load center in the current sub-area, Indicates the latitude of the load center in the current sub-area, Indicates the current sub-area The predicted load of the grid, Indicates the current sub-area The longitude of the grid, Indicates the current sub-area The latitude of the grid, Indicates the number of grids in the current sub-area.

[0071] Example 1: The total area of ​​a sub-region is 95km², the predicted load of the sub-region is 400kW, and the preset grid size is 20km². The sub-region is divided.

[0072] 1. Grid division The sub-area is divided into 4 standard grids and a residual area according to the preset grid size, and the remaining area is processed. Grid A is 20km², grid B is 20km², grid C is 20km², grid D is 20km², and the remaining area is recorded as grid E, which is 15km².

[0073] 2. Allocate forecast load The predicted load of each grid is allocated according to the grid area, so the predicted loads of grid A, grid B, grid C, and grid D are: .

[0074] The predicted load of grid E is: .

[0075] 3. Calculate the load center Assume that the geometric center coordinates of each grid (unit: km) are as follows: Grid A: ; Grid A: ; Grid C: ; Grid D: ; Grid E: .

[0076] Then calculate the load center coordinates according to the first weighted centroid formula: ; ; Therefore, the coordinates of the load center of this sub-area are , the load capacity of the load center is 400kW.

[0077] S8 architecture construction: Based on the load capacity of the load center in each sub-area, select the corresponding electrical equipment, connect the load centers in each sub-area through transmission lines, and build the power grid architecture.

[0078] S81 Select electrical equipment: Select the type and capacity of substations and transformers based on the load capacity of the load center in each sub-area.

[0079] Based on the load capacity of the load center of each sub-region, the load level of the load center of each sub-region is obtained by comparing with the preset load capacity level correspondence table, and the substation type is selected according to the load level of the load center of each sub-region. For example, for sub-regions with high load levels of load centers, large centralized substations are selected, and for sub-regions with low load levels of load centers, distributed substations or small transformer networks are selected. Substations include main transformers, standby transformers, distribution devices, and voltage control and regulation equipment.

[0080] 20% of the load capacity of the load center of each sub-area is used as the redundant load of the corresponding sub-area, and the sum of the load capacity of the load center of each sub-area and the corresponding redundant load is used as the capacity of the main transformer of each sub-area. The capacity of the standby transformer is consistent with that of the main transformer, and the standby transformer is used to provide redundant support when the main transformer fails. The distribution device includes distribution busbars, circuit breakers, switchgear, etc., which are used to distribute the power of the substation to various areas and ensure the safety and stability of the power supply. The voltage control and regulation equipment includes voltage regulators, capacitors, voltage regulating devices, etc., which are used to control the voltage within a predetermined range, ensure the stable operation of the power grid, and avoid excessively high or low voltage affecting electrical equipment and user power consumption.

[0081] S82 determines the grid topology: designs the grid topology according to the coordinates and load capacity of the load center of each sub-area.

[0082] Preset the load screening threshold, compare the load capacity of each load center with the load screening threshold, record the load center with a load capacity greater than the preset load screening threshold as the first load center, and record the load center with a load capacity less than or equal to the preset load screening threshold as the second load center. Use the trunk line to connect each first load center, use the branch line to connect each second load center with each first load center, and form a ring network structure with the trunk line and the branch line to obtain the power grid topology. At the same time, use software tools such as PSS / E, DIgSILENT and other software to simulate the load flow of the obtained power grid topology, analyze the current and voltage distribution of the entire power grid, ensure that the voltage fluctuates stably within the specified range, and avoid excessively high or low voltages that affect the safe operation of the power grid.

[0083] S9 load control: includes S91 third judgment and S92 fourth judgment.

[0084] S91 third judgment: obtain the actual charging load of each sub-area, calculate the difference between the actual charging load and the predicted load of each sub-area, record it as the first difference, and judge whether the first difference of the sub-area is higher than the preset first load threshold.

[0085] If so, the sub-region where the first difference is higher than the first load threshold is recorded as an overload region, and the fourth determination in S92 is executed.

[0086] If not, no action will be taken.

[0087] S92 Fourth judgment: record the absolute value of the first difference of the remaining sub-areas except the overload area as the second difference, and judge whether the second difference of the remaining sub-areas except the overload area is higher than the preset second load threshold.

[0088] If so, the sub-area where the second difference is higher than the second load threshold is recorded as an idle area, and the difference between the second difference of the idle area and the preset third load threshold is calculated and recorded as the allocated load. The allocated load of the idle area is allocated to the overloaded area, and the charging power of the charging piles in the overloaded area is reduced to the first preset charging power.

[0089] If not, the charging power of the charging pile in the overload area is reduced to a second preset charging power.

[0090] In the fourth judgment at S92, the allocated load of the idle area is allocated to the overloaded area, including the first sorting at S921, the second sorting at S922, the first allocation at S923, and the second allocation at S924.

[0091] S921 First sorting: sorting the overloaded areas in descending order based on the first difference values ​​of the overloaded areas, and obtaining a load difference sorting table of the overloaded areas.

[0092] S922 Second sorting: Calculate the distance between the load center of the first-ranked overloaded area and the load centers of each idle area, sort them in ascending order, and obtain an idle area distance sorting table of the first-ranked overloaded area.

[0093] S923 First allocation: preferentially select the idle area ranked first in the idle area distance ranking table to allocate load to the overloaded area. If the allocated load is still less than the first difference of the overloaded area at this time, select the idle area ranked second in the idle area distance ranking table of the overloaded area to allocate load to the overloaded area, until the total allocated load is equal to the first difference of the overloaded area.

[0094] S924 Second Allocation: According to the overload area load difference ranking table, perform S922 second sorting to S923 first allocation for each overload area in turn until the load allocation task for each overload area is completed. If the allocated load of the idle area is not enough to complete the load allocation task of all overload areas, the charging power of the charging piles in the overload area that has not completed the load allocation task is reduced to the second preset charging power.

[0095] Example: A power grid includes four sub-areas, namely area a, area b, area c and area d, and each sub-area has multiple charging piles.

[0096] The predicted load of area a is 100kWh, and the actual charging load of area a is 150kWh. The predicted load of area b is 180kWh, and the actual charging load of area b is 110kWh. The predicted load of area c is 200kWh, and the actual charging load of area c is 220kWh. The predicted load of area d is 90kWh, and the actual charging load of area d is 80kWh.

[0097] The first load threshold is 0 kWh, the second load threshold is 10 kWh, and the third load threshold is 15 kWh.

[0098] 1. According to the calculation, the first difference in area a is 50 kWh, the first difference in area b is -70 kWh, the first difference in area c is 20 kWh, and the first difference in area d is -10 kWh.

[0099] 2. Compare the first difference value of each area with the first load threshold value. The first difference value of area a and area c is greater than the first load threshold value, so area a and area c are overloaded areas.

[0100] 3. The second difference values ​​of area b and area d are calculated to be 70 kWh and 10 kWh respectively. The second difference values ​​of area b and area d are compared with the second load threshold. If the second difference value of area b is greater than the second load threshold, area b is an idle area. The difference between the second difference value of area b and the third load threshold is calculated, and the dispatched load of area b is 55 kWh.

[0101] 4. Sort area a and area c in descending order based on the first difference between area a and area c, with area a ranking first and area c ranking second. Then, reduce the charging power of the charging piles in area a and area c to the first preset charging power.

[0102] 5. Area B first allocates 50kWh of load to area a, and then allocates the remaining 5kWh of load to area c, further reducing the charging power of the charging piles in area c to the second preset charging power.

[0103] This embodiment is a power grid architecture constructed based on the charging demand of charging piles.

[0104] In this embodiment, a charging demand prediction model is constructed to predict the charging demand of each charging pile in different time periods, which can accurately predict future charging demand, identify charging peak and valley periods in advance, help to reasonably dispatch power resources, and reduce the overload of the power grid due to sudden load fluctuations. In addition, based on the location data of each charging pile, the charging demand prediction data of each time period and the regional characteristic data, the clustering algorithm is used to divide the power grid into regions, and the charging pile location and the corresponding charging demand prediction data are divided into multiple regions. It can identify the regions with high and low charging demand and the regions with similar charging demand time fluctuations, so as to allocate power resources in a targeted manner, reduce the overload and idle resources in each region, and help to plan the power grid capacity more reasonably, thereby optimizing the configuration of power resources and improving the adaptability of the power grid to load changes. Moreover, based on the divided power grid, the load center is set, and the appropriate electrical equipment and transmission lines are selected based on the set load center to construct the power grid architecture, which can accurately grasp the predicted load changes of each sub-region, adjust the power distribution in time, optimize the load distribution of the power grid, and improve the stability and safety of the power grid.

[0105] Embodiment 2: The difference from Embodiment 1 is as follows: like Figure 3As shown, after executing S3 the third acquisition and before executing S4 model building, it also includes S31 traffic flow prediction, and S31 traffic flow prediction includes S311 first area division, S312 building a traffic network diagram, S313 second area division, S314 first data matching, S315 second data matching, S316 calculation of transfer probability, S317 conditional probability optimization, S318 dynamic modeling of traffic flow and S319 acquisition of traffic flow distribution.

[0106] S311 First area division: Based on the population density and building function of each area, the K-means clustering algorithm is used to perform the first area division on each area to obtain a new area grouping result, which is recorded as the first traffic area.

[0107] S312 constructs a traffic network diagram: maps the travel habits of car owners and regional traffic flow in each first traffic area, obtains the travel habits of car owners and regional traffic flow in each first traffic area, and constructs a traffic network diagram based on the travel habits of car owners and regional traffic flow in each first traffic area, with each first traffic area as a node, and defines the connection relationship of the edges based on the travel habits of car owners in each first traffic area and the corresponding regional traffic flow.

[0108] S313 Second area division: Based on the traffic network diagram, Louvain algorithm is used to perform a second area division on each first traffic area, and the result after the second area division is used as a new first traffic area.

[0109] S314 first data matching: mapping the historical vehicle owner travel habits and the historical regional traffic flow to each first traffic area, obtaining new historical vehicle owner travel habits and new historical regional traffic flow for each first traffic area, which are recorded as first historical vehicle owner travel data and first historical traffic flow data respectively.

[0110] S315 Second data matching: Use historical holiday conditions to divide the first historical vehicle owner travel data and the first historical traffic flow data, respectively, to obtain non-holiday historical vehicle owner travel data, holiday historical vehicle owner travel data, non-holiday historical traffic flow data and holiday historical traffic flow data; record the non-holiday historical vehicle owner travel data and non-holiday historical traffic flow data as the first traffic data, and record the holiday historical vehicle owner travel data and holiday historical traffic flow data as the second traffic data.

[0111] S316 calculates the transfer probability: based on the first traffic data and the second traffic data respectively, the Markov chain method is used to calculate the first vehicle transfer probability and the second transfer probability of each first traffic area, the first vehicle transfer probability and the second transfer probability of the first traffic area are weightedly summed, and the weighted sum result is recorded as the comprehensive transfer probability of the first traffic area.

[0112] The formula for calculating the car transfer probability is: .

[0113] in, Indicates from the first traffic area To the first traffic area The car transfer probability, From the first traffic area To the first traffic area traffic flow, Indicates from the first traffic area The sum of traffic flows to all first traffic areas, is an index, indicating traversal of all first traffic areas.

[0114] S317 Conditional Probability Optimization: including S3171 fourth data collection, S3172 definition of Bayesian network, S3173 construction of conditional probability table and S3174 optimization of transition probability, such as Figure 4 shown.

[0115] S3171 Fourth data collection: collecting historical condition data corresponding to the first historical vehicle owner travel data and the first historical traffic flow data, the historical condition data including historical weather conditions and historical electricity prices.

[0116] Historical weather data can be obtained through the official website of the China Meteorological Administration, and historical electricity prices can be obtained through power companies.

[0117] S3172 defines a Bayesian network: based on the historical condition data corresponding to the travel habits of the historical car owners in each first traffic area, defines the Bayesian network variables and the causal relationship between the variables; the Bayesian network variables include child variables and parent variables. The parent variables are weather conditions and electricity prices, and the child variables are the travel habits of the car owners.

[0118] S3173 builds a conditional probability table: fully arrange the parent variable values ​​of the child variables to obtain several condition combinations (such as weather conditions: sunny, electricity prices: low electricity prices, travel habits: 0.6 frequent travel; 0.4 no travel), calculate the probability value of each condition combination based on historical condition data and historical car owners' travel habits, and collect the probability values ​​of each condition combination to obtain a conditional probability table.

[0119] S3174 optimizes the transfer probability: based on the historical condition data of each first traffic area, obtains the condition data corresponding to the first traffic data and the second traffic data, queries the condition probability table according to the obtained condition data, obtains the condition probability, uses the condition probability to optimize the comprehensive transfer probability of the first traffic area using the Bayesian method, and uses the optimized comprehensive transfer probability as the new comprehensive transfer probability.

[0120] The Bayesian formula is: .

[0121] in, Indicates that the conditional data The comprehensive transition probability under Represents conditional data The conditional probability of represents the comprehensive transition probability before optimization, Represents conditional data The total probability of =1, which is the normalization coefficient.

[0122] S318 Traffic flow dynamic modeling: Gather the comprehensive transfer probabilities of each first traffic area and build a traffic flow dynamic model.

[0123] Define the traffic flow distribution of each first traffic area at the initial time as .in, represents the traffic flow of all first traffic areas at the initial moment, Indicates the initial time Traffic flow in the first traffic area.

[0124] The comprehensive transfer probabilities of each first traffic area are collected to obtain a comprehensive transfer probability matrix, and a traffic flow dynamic model is constructed based on the comprehensive probability matrix.

[0125] .

[0126] in, Indicates time traffic flow, Indicates time The comprehensive transition probability matrix, Indicates time of traffic flow.

[0127] S3181 Fifth data collection: collect real-time condition data, which includes weather conditions and electricity prices.

[0128] S3182 data labeling: Use timestamps to match weather conditions, electricity prices, and regional traffic flows, label the regional traffic flows with weather conditions and electricity prices, and use the labeled regional traffic flows as new regional traffic flows.

[0129] S319 obtains the traffic flow distribution: inputs the regional traffic flow into the traffic flow dynamic model, obtains the predicted traffic flow distribution data of each region, and uses the predicted traffic flow distribution data as the new regional traffic flow.

[0130] In this embodiment, the predicted traffic flow distribution data is used as the new regional traffic flow, and the charging demand of each charging pile in each time period is predicted in combination with the corresponding holiday conditions.

[0131] In this embodiment, the regions are administrative regions of the region.

[0132] In this embodiment, a dynamic model of vehicle flow is constructed, and combined with the comprehensive transfer probability, the vehicle flow distribution between regions can be dynamically simulated. By predicting future traffic flow, the prediction range of the charging demand prediction model can be extended to several days or even weeks, which is suitable for charging demand planning and power grid load scheduling on a longer time scale, and helps to identify the peak charging period of charging piles in advance, and schedule the power grid load in advance, reducing the overload of local power grids due to load concentration. In addition, the introduction of traffic network graphs and graph partitioning algorithms in the process of regional division makes the regional division more reasonable, and also helps the dynamic model of vehicle flow to adapt to the characteristics of traffic networks in different cities or regions, and improves the scalability of the dynamic model of vehicle flow. Moreover, the introduction of Bayesian networks and conditional probability tables considers the impact of external factors on car owner behavior and traffic flow, improves the prediction accuracy and adaptability of the dynamic model of vehicle flow, and improves the accuracy of traffic flow prediction, which not only improves the accuracy and real-time prediction of the charging demand prediction model, but also helps the power grid and charging pile operators to dispatch resources more efficiently, optimize services, and reduce the imbalance of supply and demand caused by sudden changes in conditions, so that the power grid can more effectively cope with complex and changeable traffic environments.

[0133] Embodiment 3: The difference from Embodiment 1 is as follows: After executing S6 power grid division and before executing S7 first setting, the method further includes: S611 first judgment: determine whether there is a charging pile in the sub-area, if yes, execute S612 second judgment; if not, execute S7 first setting.

[0134] S612 Second judgment: judge whether the number of charging piles in the sub-area is greater than 1.

[0135] If so, the charging pile load weight is calculated based on the charging demand forecast data of each charging pile. According to the locations of all charging piles in the current sub-area and the corresponding load weights, the coordinates of the load center of the current sub-area are calculated by the second weighted centroid formula. The predicted load of the current sub-area is used as the load capacity of the load center, and the S8 architecture construction steps are executed.

[0136] If not, the coordinates of the charging pile are used as the load center of the current sub-area, the predicted load of the current sub-area is used as the load capacity of the load center, and the S8 architecture is constructed.

[0137] The formula for calculating the charging pile load weight is: .

[0138] in, Indicates charging pile The load weight, Indicates charging pile Charging demand forecast data, Indicates the forecast load of the current sub-area.

[0139] The second weighted centroid formula is: ; .

[0140] in, Indicates the longitude of the load center in the current sub-area, Indicates the latitude of the load center in the current sub-area, Indicates charging pile The load weight, Indicates the charging piles in the current sub-area The longitude, Indicates the charging piles in the current sub-area The latitude, Indicates the number of charging piles in the current sub-area.

[0141] In this embodiment, the location of the load center is set according to the distribution of charging piles in the sub-area, taking into account the load weight and location of each charging pile, to ensure that the location of the load center best represents the charging demand distribution in the area. This can effectively allocate grid resources, make power transmission more efficient, reduce unnecessary power waste, and improve the power supply efficiency and stability of the grid.

[0142] Embodiment 4: The difference from Embodiment 1 is as follows: After executing the step of dividing the power grid in S6 and before executing the first setting in S7, the method further includes: S621 calculates load density: obtains the area of ​​each sub-region based on the geographic data of each sub-region, calculates the load density of each sub-region in each time period based on the area of ​​each sub-region and the predicted load of each charging pile in each sub-region in each time period, aggregates the load density of each sub-region in each time period, and obtains the load density characteristic vector of each sub-region.

[0143] S622 power grid secondary division: K-means clustering algorithm is used to cluster each sub-region into different preliminary sub-regions using the load density feature vector of each sub-region, and the preliminary sub-regions are aggregated to obtain a preliminary region set.

[0144] S623 constructs a regional connection graph: based on the power grid equipment data of each region, each preliminary sub-region is used as a graph node, the transmission lines between the preliminary sub-regions are used as edges, and the edge weight is set according to the capacity and distance of the transmission lines between the preliminary sub-regions to construct a regional connection graph. The edge weight is the ratio of the capacity of the transmission line to the distance of the transmission line.

[0145] S624 Region Optimization: Based on the region connection graph, the minimum cut algorithm is used for optimization, and each optimized region is used as a new sub-region.

[0146] In the regional connection graph, any two nodes are designated as the source point and the sink point respectively. The maximum flow algorithm (such as Ford-Fulkerson algorithm) is used to calculate the maximum flow from the source point to the sink point. According to the maximum flow result, each node in the regional connection graph is divided into a source point set and a sink point set, so that the sum of the weights of the edges crossing the source point set and the sink point set is minimized, and the minimum cut result in the regional connection graph is obtained. According to the minimum cut result, the preliminary sub-region is re-divided, and the re-divided area is used as the new sub-region.

[0147] In this embodiment, the regions are administrative regions.

[0148] In this embodiment, during the grid division process, by calculating the load density characteristic vector of the sub-region, the influence of the regional area and the charging load distribution is comprehensively considered, so that the sub-region division result is more in line with the actual needs. In addition, based on the obtained preliminary regional set, the regional connection diagram is constructed, and the division of the sub-region is further optimized by the minimum cut algorithm, so that the grid can adapt to the demand changes more flexibly when facing the fluctuation of charging demand in different regions, thereby improving the adaptability of the grid. Moreover, optimization based on the load density and the structure of the transmission network is helpful to formulate a more scientific long-term development plan for the grid.

[0149] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A parameterized three-dimensional power grid intelligent construction method, characterized in that: include: First collection: collect geographical data, power grid equipment data and regional characteristic data of each region based on the geographic information system; Grid equipment data includes charging station location data; Second collection: collect the first real-time data of each area, including holiday conditions, regional traffic flow and car owners' travel habits; Third collection: collecting the first historical data of each area, the first historical data includes the historical charging data of each charging pile, the historical regional traffic flow, the historical holiday conditions and the historical travel habits of car owners; Model construction: Build a charging demand prediction model; Demand forecasting: inputting the first real-time data into the charging demand forecasting model to obtain charging demand forecasting data of each charging pile; Grid division: Cluster the charging piles based on their location data, charging demand forecast data and regional feature data. According to the clustering results, several sub-areas are obtained. The sum of the charging demand forecast data of the charging piles in each sub-area is used as the forecast load of each sub-area. The first setting is: divide the sub-areas according to the preset grid size, divide the predicted load of each sub-area based on the area of ​​each grid, obtain the predicted load of each grid, obtain the geometric center coordinates of each grid, use the weighted centroid method to calculate the coordinates of the load center of each sub-area, and use the predicted load of each sub-area as the load capacity of the load center; Architecture construction: Based on the load capacity of the load centers in each sub-region, the load centers in each sub-region are interconnected to build the power grid architecture.

2. The method for intelligently constructing a parameterized three-dimensional power grid according to claim 1, characterized in that: After executing the third acquisition step and before executing the model building step, the method further includes: First area division: Based on the regional characteristic data of each area, a clustering algorithm is used to perform the first area division on each area to obtain a new area grouping result, which is recorded as the first traffic area; First data matching: mapping the historical vehicle owner travel habits and historical regional traffic flow to each first traffic area, obtaining new historical vehicle owner travel habits and new historical regional traffic flow in each first traffic area, which are recorded as first historical vehicle owner travel data and first historical traffic flow data respectively; Second data matching: using historical holiday conditions to divide the first historical vehicle owner travel data and the first historical traffic flow data, respectively, to obtain non-holiday historical vehicle owner travel data, holiday historical vehicle owner travel data, non-holiday historical traffic flow data, and holiday historical traffic flow data, and record the non-holiday historical vehicle owner travel data and non-holiday historical traffic flow data as the first traffic data, and record the holiday historical vehicle owner travel data and holiday historical traffic flow data as the second traffic data; Calculate the transfer probability: based on the first traffic data and the second traffic data, use the Markov chain method to calculate the first vehicle transfer probability and the second vehicle transfer probability of each first traffic area, and record the first vehicle transfer probability and the second vehicle transfer probability of the first traffic area as the comprehensive transfer probability of the first traffic area; Traffic flow dynamic modeling: collect the comprehensive transfer probabilities of each first traffic area and build a traffic flow dynamic model; Obtaining traffic flow distribution: Inputting regional traffic flow into the traffic flow dynamic model, obtaining predicted traffic flow distribution data for each region, and using the predicted traffic flow distribution data as the new regional traffic flow.

3. The method for intelligently constructing a parameterized three-dimensional power grid according to claim 2, characterized in that: After performing the first area division step and before performing the first data matching step, the method further includes: Constructing a traffic network diagram: mapping the travel habits of car owners and regional traffic flow in each first traffic area, obtaining the travel habits of car owners and regional traffic flow in each first traffic area, and constructing a traffic network diagram with each first traffic area as a node and the travel habits of car owners in each first traffic area and the corresponding regional traffic flow as an edge; Second area division: Based on the traffic network graph, a graph division algorithm is used to perform a second area division on each first traffic area, and the result of the second area division is used as the new first traffic area.

4. The method for intelligently constructing a parameterized three-dimensional power grid according to claim 2, characterized in that: After executing the step of calculating the transfer probability and before executing the step of dynamic modeling of traffic flow, the method further includes: Fourth data collection: collecting historical condition data corresponding to the first historical vehicle owner travel data and the first historical traffic flow data, the historical condition data including historical weather conditions and historical electricity prices; Defining a Bayesian network: defining Bayesian network variables and causal relationships between variables based on the historical travel habits of vehicle owners in each first traffic area and the corresponding historical condition data; the Bayesian network variables include child variables and parent variables; Construct a conditional probability table: fully arrange the parent variable values ​​of the child variables to obtain several condition combinations, calculate the probability value of each condition combination based on historical condition data and historical vehicle owner travel habits, and collect the probability values ​​of each condition combination to obtain a conditional probability table; Optimize the transfer probability: based on the historical condition data of each first traffic area, obtain the condition data corresponding to the first traffic data and the second traffic data, query the condition probability table according to the obtained condition data, obtain the condition probability, use the condition probability to optimize the comprehensive transfer probability of the first traffic area, and use the optimized comprehensive transfer probability as the new comprehensive transfer probability; After executing the step of dynamic modeling of vehicle flow, and before executing the step of obtaining vehicle flow distribution, the method further includes: Fifth, data collection: collect real-time condition data, including weather conditions and electricity prices; Data labeling: Use timestamps to match weather conditions, electricity prices, and regional traffic flows, label the regional traffic flows with weather conditions and electricity prices, and use the labeled regional traffic flows as new regional traffic flows.

5. The method for intelligently constructing a parameterized three-dimensional power grid according to claim 1, characterized in that: After the model building step and before the demand forecasting step, the following steps are also included: Constructing a sample training set: constructing a sample training set based on the first historical data; Encoder training: The sample training set is input into the charging demand prediction model, and the Transformer encoder encodes the sample data in the sample training set; Decoder prediction: The Transformer decoder decodes and predicts the encoded sample data to obtain charging demand prediction data; Iterative optimization: define a mean square error loss function, iteratively optimize the charging demand prediction model by minimizing the mean square error loss function, and use the optimized charging demand prediction model as the new charging demand prediction model.

6. The method for intelligently constructing a parameterized three-dimensional power grid according to claim 1, characterized in that: After executing the step of dividing the power grid and before executing the step of first setting, the method further includes: First judgment: determine whether there is a charging pile in the sub-area: If yes, then execute the second judgment step; If not, then execute the first setting step; Second judgment: judge whether the number of charging piles in the sub-area is greater than 1: If yes, the load weight of each charging pile is calculated according to the charging demand forecast data of each charging pile, the coordinates of the load center of the current sub-area are calculated by the weighted centroid method according to the positions of all charging piles in the current sub-area and the corresponding load weights, the predicted load of the current sub-area is used as the load capacity of the load center, and the step of architecture construction is executed; If not, the coordinates of the charging pile are used as the load center of the current sub-area, the predicted load of the current sub-area is used as the load capacity of the load center, and the architecture construction step is executed.

7. The method for intelligently constructing a parameterized three-dimensional power grid according to claim 1, characterized in that: After executing the step of dividing the power grid and before executing the step of first setting, the method further includes: Calculate load density: According to the geographical data of each sub-region, obtain the area of ​​each sub-region, calculate the load density of each sub-region in each period according to the area of ​​each sub-region and the predicted load of each charging pile in each sub-region in each period, aggregate the load density of each sub-region in each period, and obtain the load density characteristic vector of each sub-region; Secondary division of power grid: clustering algorithm is used to cluster each sub-region into different preliminary sub-regions using the load density feature vector of each sub-region, and the preliminary sub-regions are aggregated to obtain a preliminary region set; Constructing a regional connection graph: Based on the power grid equipment data of each region, each preliminary sub-region is used as a graph node, and the transmission lines between the preliminary sub-regions are used as edges to construct a regional connection graph; Region optimization: Based on the region connection graph, the minimum cut algorithm is used for optimization, and the optimized preliminary sub-regions are used as new sub-regions.

8. The method for intelligently constructing a parameterized three-dimensional power grid according to claim 7, characterized in that: After executing the architecture building steps, it also includes: Third judgment: obtain the actual charging load of each sub-area, calculate the difference between the actual charging load and the predicted load of each sub-area, record it as the first difference, and judge whether the first difference of the sub-area is higher than the preset first load threshold: If yes, the sub-region where the first difference is higher than the first load threshold is recorded as an overload region, and the fourth determination step is executed; If not, no action will be taken; Fourth judgment: record the absolute value of the first difference of the remaining sub-areas except the overload area as the second difference, and judge whether the second difference of the remaining sub-areas except the overload area is higher than the preset second load threshold: If yes, the sub-area where the second difference is higher than the second load threshold is recorded as an idle area, the difference between the second difference of the idle area and the preset third load threshold is calculated and recorded as the deployed load, the deployed load of the idle area is deployed to the overloaded area, and the charging power of the charging piles in the overloaded area is reduced; If not, the charging power of the charging piles in the overload area is reduced.

Images